This project investigates the physical parameters of the electroconvulsive therapy stimulus. A comparison trial using a standard commercial AC instrument producing 150 volts RMS for .5 seconds (energy 35-40 watt seconds) and an experimental machine producing square pulse 1 msec wide, 10 msec interpulse interval, alternate positive and negative pulses, 150 pulse train, 200 volts at 600 ma (energy 18-21 watt seconds) demonstrated equivalent clinical efficacy as measured on several scales and check lists. The experimental treatment was not only significantly lower in energy but of much greater reliability in eliciting a convulsive response. Attempts to measure possible differences in induced dysfunction were unsuccessful since most subjects showed severe cognitive deficits prior to ECT. A threshold study of the number of pulses factor showed complete reliability above 130 pulses with other parameter values as listed above. Individual limens ranged down to a minimum of 60 pulses with marked inter-individual variabiliy. A tendency in some subjects for an increase in threshold as a function of the number of treatments was established. Further efforts will be directed towards establishing threshold values of voltage and current and a trial of bilateral ECT using the brief stimulus method. BIBLIOGRAPHIC REFERENCE: Williams, R., Weaver, L., Rush, S., and Smith D. Application of a muscle-potential monitor to electroconvulsive therapy. (Conditionally accepted by Institute of Electrical and Electronic Engineering Transactions, Biomedical Engineering Section, August, 1975).